Linguistic Modeling and its Interfaces
Talk by Patrick Rebuschat und Simón Ruiz, 26 July
LEAD's Distinguished International Professor Patrick Rebuschat (Lancaster University, UK) and LEAD graduate Simón Ruiz will hold a talk presenting joint work with Padraic Monaghan (University of Amsterdam):
"The Role of Exposure Condition on the Cross Situational Learning of Vocabulary and Morphosyntax"
Friday, July 26, 10:15AM, Blochbau 1.13 (Wilhelmstr. 19)
ABSTRACT. The use of mixed-effects regression modeling as a tool for statistical analysis of second language data has been increasingly recommended in recent years (see, e.g., Cunnings, 2012; Cunnings & Finlayson, 2015; Linck & Cunnings, 2015; Murakami, 2016; Plonsky, 2017; Plonsky & Oswald, 2017) because it enables variability across individuals and across particular stimuli to be included in the model. Thus, design effects as well as inter-individual differences can be revealed within a single analysis. However, a less often exploited property of these analyses is that the more immediate context in which a stimulus is processed can also be included, such as which structure has just been experienced by the individual and its effect on current learning.
In this paper, we illustrate the potential utility of this form of analysis by presenting an experimental study on the second language acquisition of vocabulary and morphosyntax. In this artificial language study, 90 participants learned nouns, verbs, adjectives, morphological markers, and word order by keeping track of cross-situational statistics. During learning, participants viewed two scenes each comprising two aliens performing a transitive action presented on a computer. Participants simultaneously heard a sentence, composed of two nouns, a verb, two marker words indicating which noun was the subject and which the object, and optional adjectives describing the aliens’ colour. Word order varied across SOV and OSV. One of the scenes was described by the sentence, and over training participants were required to learn co-occurrences between the sentence and the target scene (through sensitivity to relations between particular words and their referents). During testing, knowledge of words and of word order was individuated. We manipulated global properties of learning, by varying exposure across three conditions. An implicit condition provided no information about the language structure and no feedback about performance. An explicit condition provided information about word order and the presence of the marker words. Finally, a feedback condition provided no information about the language structure but feedback in the form of a ringing bell played by the computer when participants correctly selected the target scene. Thus, this study extends previous studies of explicit versus implicit learning of grammar (e.g., Krashen, 1982; Nazari, 2013) by testing both grammatical and vocabulary learning, and by investigating the additional feedback condition, which provides information to the participant about how their learning is proceeding, but without explicit information about the structure to be acquired.
We found that learning was effective in all conditions. However, the feedback condition resulted in better learning of vocabulary than both implicit and explicit instruction conditions, which did not differ. In contrast, word order syntax was better learned in the implicit and explicit conditions compared to the feedback condition. Thus global properties of learning affected acquisition of vocabulary and syntax, with performance on each aspect of learning boosted by distinct instruction conditions. To acquire vocabulary, information about learning performance was most effective, but to acquire syntax, this feedback on performance was slightly detrimental. Focusing only on instructions for grammatical learning (e.g., Nazari, 2013) thus limits our understanding of how language learning proceeds. In terms of local properties of learning, we analysed whether the learner’s performance on the previous trial containing each piece of information – the nouns, the verb, and the adjectives – affected accuracy on the current learning trial. We found that the precise context of the participants’ learning of verbs and adjectives, but not nouns, influenced current performance. These results show the importance not only of global instructions, but also contingently taking into account the learner’s recent learning experience in determining effectiveness of language learning. Our paradigm thus illustrates how both vocabulary and syntax can be learned in the laboratory, and using GLME analysis, we show how learning different aspects of language are affected differently by instruction and by context of learning.Back